Evaluation of a surrogate virus neutralization assay for detecting neutralizing antibodies against SARS-CoV-2 in an African population

Abstract

The global resurgence of coronaviruses and the move to incorporate COVID-19 vaccines into the expanded program for immunization have warranted for a high-throughput and low-cost assay to measure and quantify mounted neutralizing antibodies as an indicator for protection against SARS-CoV-2. Hence, we evaluated the surrogate-virus-neutralization-assay (sVNT) as an alternative assay to the pseudo-virus neutralization assay (pVNT). The sVNT was used to measure neutralizing antibodies among 119 infected and/or vaccinated blood samples, against wild-type SARS-CoV-2 (WT) and the Omicron-variant with reference to the pVNT. Four different cut-offs were assessed for suitability in distinguishing neutralizers: the manufacturer (>30%), literature-based (>50%) and (>80%), and population-based (>27.69%). The obtained data was analyzed using "R" through its integrated development environments; JAMOV and R-Studio. Using the WT strain, only the population-based cut-off was able to differentiate neutralizers from non-neutralizers beyond chance, with an area under the curve (AUC) of 0.833 (95%CI, 0.505-1.0; <i>P</i> = .049). Applying the population-based cut-off, improved the sensitivity to 100% from 91.4% obtained from the manufacturer cut-off (<i>P</i> = .002). However, the specificity remained low (67%). The negative-predictive-value also improved to 100% vs 16.4% (<i>P</i> = .006), but there was no difference in the positive-predictive-value (99.1% vs 99.1%) (<i>P</i> = .340). When we used the Omicron-variant, the sVNT titers were not able to predict the neutralizers and non-neutralizers with reference to pVNT (AUC of 0.649) (<i>P</i> = .221). The sVNT assay is a potential alternative for screening individuals harboring potent neutralizing antibody with high sensitivity, although we recommend continuous improvement of the assay in line with the viral mutations. Further, we recommend that individual users establish a population-based cut-off while using the sVNT assay.